Editorial standards

We aim to provide members of the public access to accurate information and empower them to make informed decisions. The standards set out below reflect the values of editorial independence, impartiality, and ethical journalism that editors must abide by to maintain public trust in our work.

  1. Editors are expected to evaluate claims in an accurate and impartial manner, free of bias or external influence. They must verify that the information they use comes from trustworthy sources and provide citations for supporting evidence that they use in a review. They should use the best available evidence coming from primary sources (original works like scientific articles published in top-tier peer-reviewed journals) wherever suitable primary sources are available. If primary sources are not available, the editor should explain the use of a secondary source. They should present information with sufficient context and avoid cherry-picking.
  2. Editors should provide at least two, though preferably more, sources to verify the central claim of an investigation except in cases where there is only a single relevant source. They should avoid exaggeration (e.g. presenting the findings of a single preprint as conclusive evidence of a scientific phenomenon or process). If there is reasonable doubt or uncertainty by the scientific community over an issue, it is the editor’s duty to inform the reader about this. On the other hand, editors should also avoid introducing uncertainty or conflicting opinions into a review solely for the sake of appearing “balanced” (false balance). Being unbiased requires us to consider all the evidence at hand and give them due weight according to their quality, regardless of where they come from. It does not mean that we need to give factually incorrect or unsubstantiated views equal weight or airtime as facts.
  3. When quoting a person or a publication, the quote should ideally be presented in full. In cases where a quote needs to be truncated for the sake of brevity, the editor must ensure that they provide sufficient context to understand the quote accurately and avoid changes that distort the original meaning. Alternatively, if the quote requires heavy editing (e.g. due to errors in spelling and grammar, wordiness), the editor can instead paraphrase.
  4. Any errors that are brought to an editor’s attention must be corrected quickly and transparently. If a significant error occurs in which the original verdict of a review must be changed, a correction notice in the review should explain how the error occurred, and the correction must be communicated through all channels in which the original erroneous review was published (e.g. Twitter, Facebook).
  5. Where appropriate and possible, the editor should reach out to the person who made the claim. This is required when there is reasonable doubt over the interpretation of the claim, or when the person is subject to significant criticism. The editor is not obliged to contact the source of the claim in the following cases:
    1. the claim is one that has been shown to be false by the preponderance of scientific evidence (e.g. “vaccines cause autism”, “the Earth is flat”, “the Sun revolves around the Earth”, “humans are not the main cause of current climate change”);
    2. the source repeats a claim that was previously reviewed and found to be inaccurate;
    3. if it is likely to result in harassment or abuse of our staff.
  6. Minor cosmetic changes to a review, such as edits to punctuation, do not require an update note. If a review is updated to contain new information post-publication, an update note must be added to the review, stating the date on which the addition was made, where the addition was made in the review, and a summary of the newly added information.
  7. When quoting scientists, editors should establish the scientists’ credibility by providing information on their area of expertise in the review (e.g. by linking to the scientist’s academic or professional profile/list of scientific publications like ResearchGate or LinkedIn). Their area of expertise should be compatible with the particular subject under review. For example, while both epidemiologists and infectious disease experts broadly work in the field of public health, these are two very different specialties that involve very different subject knowledge. Editors should consult the one whose expertise is most closely related to the review.
  8. Editors must protect the privacy of individuals who may be implicated in a claim or the propagation of a claim. It is acceptable to name public figures and individuals with a significant following, including but not limited to politicians, celebrities, and social media influencers, but not for an otherwise ordinary social media user, as this leaves them vulnerable to harassment and bullying. In addition, personal identifiable information, such as addresses and phone numbers should not be published. If such information is present in an image, it should be anonymized by blurring or obscuring it. If publishing this information is necessary to debunk a claim, the information should be partially obscured (e.g. blurring or obscuring part of a phone number). Editors must also take special care to protect the privacy of minors who may be implicated in the review of a claim. This includes both minors who may be the subject of a claim or who are key sharers of misinformation.
  9. As a signatory of the International Fact-Checking Network’s Code of Principles, Science Feedback is committed to non-partisanship. In assessing the credibility of a claim, editors should be guided by the facts at hand, and not permit a personal preference for one side over another to influence their judgment.
  10. Editors may not hold a salaried or significant position in the government or in a political party, nor in public companies controlled by the aforementioned. Editors may not make statements endorsing a political party or candidate in the name of Science Feedback. This does not prevent editors from exercising their right to freedom of speech and expression in their personal capacity.
  11. If there is potential for a conflict of interest to arise during an assessment of a claim (e.g. a scientist who is part of a vaccine’s development team reviewing a claim about that vaccine), this should be disclosed clearly in the review. It may also be the case that the editor in charge of writing the review has a conflict of interest (e.g. past or previous personal or professional relationship with the claimant). In this case, the editor must voluntarily disclose this information to the editorial team. The editor may be asked to recuse themselves from the production of the review to ensure the integrity and objectivity of the review, if this is deemed necessary by the editorial team.
  12. Editors who maintain social media profiles that identify them as a member of Science Feedback must clearly state on their profiles that opinions expressed are their own. At the same time, they must also be cognizant of the fact that any content they publish on social media can influence the public’s opinion on their ability to perform their work in an unbiased manner. As such, editors must take care not to publish content, including social media comments, that call into question their ability to assess claims in an unbiased manner, or cast doubt over Science Feedback’s reputation for impartiality.
  13. Except for token items of nominal value, editors should not accept gifts from claimants and contributors to a review. This includes cash, travel, accommodation, hospitality, and entertainment. Editors may accept media passes and tickets to events if attending the event is necessary in the course of their work.

These standards are inspired by Agence France-Presse’s Editorial Standards and Best Practices, the International Fact-Checking Network’s Code of Principles, and the European Fact-Checking Standards Network’s Code of Practice.